Pre-registration

Week 6Due 2 April 2027Worth 20%

Two pages, handed in before you start. That ordering is the entire point, and it is the only deadline in this course that cannot be extended after the fact: a pre-registration submitted once you have seen your data is not a pre-registration, whatever it says on it.

What you are planning is one change to one block of your own day, and the authority for the change is public guidance rather than your own preference. Nothing here is run this week. You run it from here to week 10 and report it then, and in week 12 you run part of it again and account for the difference.

What goes in it#

1. The block. One block from your week 1 day, taken from the ones you marked as a block you would be willing to run differently and could actually reach. That mark was a note to yourself for later. This is later. If you kept the day on paper, or marked nothing, choose the block now and say that you chose it now.
2. The guideline. The published guidance that covers that block, named as a document rather than as a thing everybody knows. Week 1’s lecture is where to start looking, and the map at the foot of it is there so you can see what shapes these documents come in before you go hunting for one.
3. The four questions, from week 2’s lecture, in the order it asks them. Is it still in force. Who is it written for. Is it a recommendation or a tip. How sure is it. That order is not decoration: they are ordered by how much each one can save you, cheapest first, so that the expensive question is only ever asked of a sentence that has already survived the cheap ones. Where the document does not answer, write down that it does not answer. “This page carries no certainty rating and holds the evidence in a separate report” is a finding, and it is marked as one.
4. The change. One sentence for what you will do differently in that block, and one for which part of the guidance it comes from. Where the guidance stops short of the thing you are changing, say where it stops and that the rest is yours. Guidance narrows to a group and then runs out, and every reader of it has to close that last gap themselves.
5. The order. The changed block has to be compared against something, and the something is the same block left alone, so say here which days run changed and which run as they are. Fix that pattern now rather than deciding it each morning. If every changed day comes first and every unchanged day second, “the second stretch was better” and “I had got the hang of it by then” arrive as the same sentence, and counterbalancing is what separates them: alternate, so that being second happens equally often to both. Some changes will not alternate, because what you did on Monday is still in you on Tuesday. Caffeine, bedtime and anything else that accumulates cannot be swapped day by day, and a day by day swap would measure the accumulation rather than the change. If yours is one of those, say so, say how you can tell, and use week 5’s withdrawal design instead: a stretch as you are, a stretch changed, a stretch back as you were, with the length of each one fixed here.
6. The measure, operationalised as in week 2’s lab: what you will write down, when, and how long one recording takes. Then the part week 3 was for. State the smallest difference this measure could detect, and where that figure came from. Your two instruments disagreed by some amount on one night in week 3, and a difference smaller than that disagreement is one you have already shown yourself you cannot see. If the measure is new this week, say how you will find its spread before you rely on it.
7. The stopping rule. When you stop, and what would make you stop earlier than that. Fixed now, in writing, the way week 5 fixed the end of its removal phase before the removal began. Every week of this course above the lowest risk publishes one before it asks you to do anything, and a plan to change your own day owes you the same.
8. What would prove you wrong. A specific result, a number, a direction or a pattern, that you would accept as the change having failed.

One worked trace#

The block is two hours at a desk in the evening. The guidance is Australia’s 24-hour movement guidelines, on the page written for adults aged 18 to 64, which recommend limiting and breaking up sedentary time. The four questions: the page prints the date it was last updated, so somebody has looked at it recently; the adult page is the one addressed to a reader of that age, and there are separate pages for readers who are not; breaking up sedentary time sits among the recommendations and not under the heading marked Tips; and the page carries no certainty rating and no strength rating at all, with the supporting evidence held in a separate final report. So the sentence is in force, it is addressed to you, it is a recommendation rather than a tip, and how sure it is, this page will not say.

What the guidance does not tell you is how long a break is, where in a Thursday evening to put one, or which evenings to run changed and which to leave alone. That is the part you are writing, and saying so in the document is worth more than quietly letting the guideline appear to have covered it. Here the ordering is easy, because the block is at that desk most weeknights and standing up for two minutes is over by the time you sit back down: the two conditions alternate night by night, and neither of them is always the second one you tried.

Why the last one carries the most marks#

Because it is the one that cannot be written afterwards. Everything else in the list can be reconstructed from memory once you know how things turned out, and reconstructed honestly, with no intent to mislead. The block was in your week 1 record, the guideline is still published, the measure is in your hand, the order is a pattern of dates anyone can read off the record, and the stopping rule is a decision you could describe truthfully in October. What cannot be reconstructed is what you would have counted as failure before you knew. Writing it down in advance is the only way that sentence means anything, and it is why this piece is due before the fortnight rather than after it.

A note on what is being marked#

Your expectation does not have to be correct: a comprehensively wrong prediction can score full marks. You are marked on whether the document was specific enough to be wrong at all. Nobody reading it knows how your fortnight went, and nothing in the mark waits to find out, because marking the outcome would hand the marks to whoever happened to start on a bad week and return to an ordinary one. That is arithmetic rather than inference, and week 9 is where you watch it happen to somebody who thought it was a result.